Building structure design system based on computer aided design

Abstract

The invention relates to the technical field of computer-aided design and building engineering, in particular to a building structure design system based on computer-aided design, which comprises an initial data acquisition module for acquiring an initial structure topology model and preset load data to construct an initial force transfer diagram model; the high-risk node locking module is used for performing node failure simulation, evaluating node importance through a heuristic dimension reduction algorithm and locking a high-risk node set; the nonlinear response analysis module is used for eliminating high-risk nodes and calculating nonlinear stress distribution data and dynamic boundary conditions of residual topological structures; the redundant path optimization module is used for performing multi-path optimization under constraint to generate a secondary force transmission topology network; the toughness evaluation and optimization module is used for calculating toughness indexes and comparing threshold values, outputting if the toughness indexes reach the standard, and adjusting constraints and re-optimizing if the toughness indexes do not reach the standard; according to the method, the force transmission path can be quickly reconstructed when the structure is locally damaged, and high adaptability and robustness are shown.

Description

Technical Field

[0001] This invention relates to the field of computer-aided design and architectural engineering technology, specifically to a building structure design system based on computer-aided design. Background Technology

[0002] As the core component of engineering construction, building structures are continuously expanding in scale and becoming increasingly complex in spatial topology. To ensure the stability and safety of building systems and effectively address the risk of successive collapses under extreme conditions, conventional mechanical analysis and design are conducted on building structures based on the principle of pursuing the optimal solution for a single stiffness under static loads. However, when assessing successive collapses of building structures with complex topologies, it is necessary to calculate the mechanical response under component failure combinations. Mechanical response is a key indicator for measuring the force transmission state after local structural damage, and it directly relates to the rationality and effectiveness of redundant design. To balance the characteristics of building geometric models and stress analysis, traditional linear analysis is usually used to simulate stress distribution under static loads. The calculation of stress distribution is usually based on the assumption of small deformations. This calculation method combines geometric modeling and linear mechanical analysis, but it faces the problem of computational power explosion caused by massive node combinations when dealing with extreme conditions. This means that the real nonlinear stress mutation of the building structure after local failure is easily interfered with by traditional linear assumptions, leading to inaccurate calculation results. This limitation not only affects the rationality of topology optimization but may also negatively impact the financial feasibility of actual projects by increasing material stacking without limit. Summary of the Invention

[0003] The purpose of this invention is to provide a computer-aided design system for building structures, addressing the following technical problems: Existing conventional building structure mechanics analysis and design techniques suffer from significant shortcomings in addressing the computational burden caused by massive node combinations in assessing successive collapses under extreme conditions, inaccurate calculations of true nonlinear stress due to assumptions of minute deformations, and the potential negative impact on the financial feasibility of practical projects due to unrestricted material accumulation. There is an urgent need for a computer-aided design system that can solve the problem of exponentially increasing computational complexity through heuristic dimensionality reduction algorithms, accurately capture the true mechanical abrupt changes after local structural failure, and automatically deduce a topological network with multi-level redundancy under strict constraints on overall material consumption. This invention's objective can be achieved through the following technical solutions:

[0004] The initial data acquisition module is used to acquire the initial structural topology model and the preset load data, and to construct an initial force transmission diagram model based on the initial structural topology model and the preset load data;

[0005] The high-risk node locking module is used to simulate node failure in the initial force transmission diagram model, evaluate the importance of each node in the initial force transmission diagram model through a heuristic dimensionality reduction algorithm, and lock the set of high-risk nodes based on the importance.

[0006] The nonlinear response analysis module is used to select the nodes in the high-risk node set as target nodes and remove them from the initial force transmission diagram model, and to calculate the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target nodes.

[0007] The redundant path optimization module is used to perform multi-path optimization based on the nonlinear stress distribution data and the dynamic boundary conditions, under the preset material consumption threshold and three-dimensional geometric interference penalty constraints, to generate a secondary force transmission topology network.

[0008] The resilience assessment and optimization module is used to calculate the load redistribution response rate, topology redundancy index, and material resilience benefit ratio of the secondary force transmission topology network; compare the load redistribution response rate, topology redundancy index, and material resilience benefit ratio with preset response rate thresholds, preset redundancy thresholds, and preset benefit ratio thresholds, respectively; if all are greater than or equal to the corresponding preset thresholds, the secondary force transmission topology network is output as the target structural topology network; otherwise, the three-dimensional geometric interference penalty constraint or the preset material consumption threshold is adjusted, and the redundant path optimization module is triggered to re-perform multi-path optimization.

[0009] Furthermore, the method for constructing an initial force transmission diagram model based on the initial structural topology model and the preset load data includes:

[0010] Extract the component connection relationships from the initial structural topology model to generate a basic adjacency matrix;

[0011] Extract the component geometric parameters and material physical parameters from the initial structural topology model, and map the component geometric parameters and material physical parameters to the corresponding edges of the basic adjacency matrix to generate a weighted adjacency matrix;

[0012] The preset load data is converted into nodal force vectors;

[0013] The weighted adjacency matrix and the node force vector are fused to construct the initial force transmission graph model.

[0014] Furthermore, the method for evaluating the importance of each node in the initial force graph model using a heuristic dimensionality reduction algorithm, and for identifying a set of high-risk nodes based on the importance, includes:

[0015] Traverse all nodes in the initial force transmission diagram model to generate a node failure combination space;

[0016] The dimensionality of the node failure combination space is reduced based on preset node degree and connectivity pruning rules to obtain a subset of candidate failure nodes.

[0017] Each node in the candidate failure node subset is sequentially set to a failure state, and the global stiffness attenuation value of the initial force transmission diagram model is calculated each time it is set to a failure state.

[0018] The global stiffness attenuation value is used as the importance of the corresponding node;

[0019] Nodes with an importance greater than a preset importance threshold are extracted and combined into the high-risk node set.

[0020] Furthermore, the method for calculating the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target node includes:

[0021] The target node and the associated edges connected to the target node are deleted from the initial force transmission graph model to obtain the remaining topology.

[0022] The load is redistributed to the remaining topology to obtain the initial unbalanced force vector;

[0023] Based on the initial unbalanced force vector, an incremental iterative algorithm is used to perform large deformation analysis on the remaining topology to obtain the deformation displacement data of each remaining component in the remaining topology.

[0024] The spatial coordinates of the remaining topology are updated based on the deformation displacement data to generate the dynamic boundary conditions.

[0025] Based on the dynamic boundary conditions and the deformation displacement data, the internal force values ​​of each remaining component are calculated, and the internal force values ​​are used as the nonlinear stress distribution data.

[0026] Furthermore, methods for generating secondary force transmission topology networks include:

[0027] Using nodes in the nonlinear stress distribution data whose stress values ​​are greater than a preset stress threshold as source nodes and nodes whose stress values ​​are less than a preset safety threshold as sink nodes, an optimization space is constructed in the remaining topology.

[0028] A preset number of virtual force transmission paths are initialized in the optimization space, and each virtual force transmission path is assigned an initial flow rate.

[0029] Calculate the three-dimensional spatial coordinates of each virtual force transmission path, and perform collision detection between the three-dimensional spatial coordinates and the preset building functional space model;

[0030] If a collision occurs, the three-dimensional geometric interference penalty constraint is applied to the virtual force transmission path where the collision occurs, reducing the flow conduction rate of the corresponding virtual force transmission path; if no collision occurs, the flow conduction rate of the corresponding virtual force transmission path is maintained.

[0031] Based on the dynamic boundary conditions, the flow conduction rate, and the preset material consumption threshold, the flow distribution of all virtual force transmission paths is dynamically updated, and virtual force transmission paths with flow rates greater than the preset flow rate threshold are retained to generate the secondary force transmission topology network.

[0032] Furthermore, the method for calculating the load redistribution response rate of the secondary force transmission topology network includes:

[0033] In the secondary force transmission topology network, the amount of transferred load from the source node to the sink node is obtained;

[0034] Obtain the original load that the target node was subjected to before it was removed;

[0035] Calculate the ratio of the transferred load to the original load to obtain the foundation diversion ratio;

[0036] Record the response time consumed by the redundant path optimization module in generating the secondary force transmission topology network;

[0037] The response time is transformed based on a preset time penalty function that monotonically decreases with the response time to obtain a time decay coefficient;

[0038] The load redistribution response rate is obtained by multiplying the basic diversion ratio by the time decay coefficient.

[0039] Furthermore, the method for calculating the topology redundancy index of the secondary force transmission topology network includes:

[0040] Extract the total number of independent force transmission paths in the secondary force transmission topology network;

[0041] Calculate the degree distribution variance of all nodes in the secondary force transmission topology network;

[0042] Calculate the reciprocal of the variance of the degree distribution to obtain the distribution uniformity;

[0043] The total number of independent force transmission paths and the distribution uniformity are assigned preset weights and then summed to obtain the redundancy index of the topology network.

[0044] Furthermore, the method for calculating the consumable resilience benefit ratio of the secondary force transmission topology network includes:

[0045] Calculate the total material volume of the secondary force transmission topology network and obtain the initial material volume of the initial force transmission diagram model;

[0046] Calculate the difference between the total material volume and the initial material volume to obtain the material increment;

[0047] Perform ultimate bearing capacity analysis on the secondary force transmission topology network to obtain the network's ultimate bearing capacity;

[0048] Obtain the initial ultimate bearing capacity of the initial force transmission diagram model;

[0049] Calculate the difference between the network's ultimate bearing capacity and the initial ultimate bearing capacity to obtain the increase in bearing capacity;

[0050] Divide the increase in load-bearing capacity by the increase in material quantity to obtain the toughness benefit ratio of the consumable.

[0051] Furthermore, the method of adjusting the three-dimensional geometric interference penalty constraint or the preset material consumption threshold, and triggering the redundant path optimization module to re-perform multi-path optimization, includes:

[0052] If the load redistribution response rate is less than the preset response rate threshold, a first adjustment coefficient less than one is obtained, the constraint coefficient corresponding to the three-dimensional geometric interference penalty constraint is multiplied by the first adjustment coefficient for relaxation processing, and the redundant path optimization module is triggered.

[0053] If the load redistribution response rate is greater than or equal to the preset response rate threshold, and the topology network redundancy index is less than the preset redundancy threshold, then a second adjustment coefficient greater than one is obtained, the constraint coefficient corresponding to the three-dimensional geometric interference penalty constraint is multiplied by the second adjustment coefficient for tightening, and the redundant path optimization module is triggered.

[0054] If the load redistribution response rate is greater than or equal to the preset response rate threshold, the topology network redundancy index is greater than or equal to the preset redundancy threshold, and the material resilience benefit ratio is less than the preset benefit ratio threshold, then a preset material reduction step size is obtained, the preset material consumption threshold is reduced based on the preset material reduction step size, and the redundant path optimization module is triggered.

[0055] Compared with existing technologies, the system of this invention addresses the computational complexity problem caused by massive node combinations and the analysis error caused by the assumption of small deformations in the background technology. This system compresses the failure combination space through a heuristic dimensionality reduction algorithm, effectively avoiding exponential computing power consumption and providing high-speed response capability for large-scale building models. The system adopts an incremental iterative algorithm for large deformation nonlinear response analysis, breaking the linear premise of traditional static load analysis. It accurately captures the real mechanical mutation and nonlinear stress redistribution state at the moment of local structural failure or fracture, thus providing a high-fidelity mechanical benchmark for subsequent redundant path design. Attached Figure Description

[0056] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0057] Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0059] Example 1:

[0060] Please see Figure 1 Computer-aided design-based building structure design systems include:

[0061] The initial data acquisition module is used to acquire the initial structural topology model and preset load data, and to construct the initial force transmission diagram model based on the initial structural topology model and preset load data;

[0062] The high-risk node locking module is used to simulate node failure in the initial force transmission diagram model. It evaluates the importance of each node in the initial force transmission diagram model through a heuristic dimensionality reduction algorithm and locks the set of high-risk nodes based on the importance.

[0063] The nonlinear response analysis module is used to select nodes in the high-risk node set as target nodes and remove them from the initial force transmission diagram model, and calculate the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target nodes.

[0064] The redundant path optimization module is used to perform multi-path optimization based on nonlinear stress distribution data and dynamic boundary conditions, under the constraints of preset material consumption threshold and three-dimensional geometric interference penalty, and generate a secondary force transmission topology network.

[0065] The resilience assessment and optimization module is used to calculate the load redistribution response rate, topology redundancy index, and material resilience benefit ratio of the secondary force transmission topology network. The load redistribution response rate, topology redundancy index, and material resilience benefit ratio are compared with preset response rate thresholds, preset redundancy thresholds, and preset benefit ratio thresholds, respectively. If all are greater than or equal to the corresponding preset thresholds, the secondary force transmission topology network is output as the target structural topology network. Otherwise, the three-dimensional geometric interference penalty constraint or the preset material consumption threshold is adjusted, and the redundancy path optimization module is triggered to re-perform multi-path optimization.

[0066] This embodiment provides a computer-aided design system for building structures, breaking through the conventional design thinking of pursuing the optimal solution of a single stiffness under traditional static loads. It introduces the multi-morphic toughness routing mechanism of slime mold foraging networks in bionics. The system transforms the conventional building geometric model into a low-level data structure that can be computed by a graph neural network through an initial data acquisition module, acquires the initial structural topology model and preset load data, and constructs an initial force transmission graph model. This initial force transmission graph model abstracts building components as edges and component connection points as nodes, establishing the benchmark for weighted dynamic mathematical graphs.

[0067] The high-risk node locking module simulates node failure in the initial force transmission diagram model, aiming to solve the problem of exponentially increasing computational complexity caused by the massive component failure combinations in continuous collapse scenarios. It evaluates the importance of each node through a heuristic dimensionality reduction algorithm and locks the set of high-risk nodes, providing precise targets for subsequent redundancy design. The nonlinear response analysis module removes nodes from the set of high-risk nodes as target nodes and calculates the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology. This step breaks the assumption of small deformations in traditional linear analysis and accurately captures the real mechanical abrupt state after local structural failure.

[0068] Based on this, the redundant path optimization module performs multi-path optimization within limited physical space and material cost, based on nonlinear stress distribution data and dynamic boundary conditions, to generate a secondary force transmission topology network with self-healing capability at breakpoints.

[0069] The resilience assessment and optimization module calculates various indicators of the secondary force transmission topology network. When the load redistribution response rate, topology network redundancy index, and material consumption resilience benefit ratio are all greater than or equal to the corresponding preset thresholds, the system determines that the currently generated network balances transmission efficiency and fault tolerance, and outputs the secondary force transmission topology network as the target structural topology network. When any indicator fails to meet the standard, the system will adjust the three-dimensional geometric interference penalty constraint or the preset material consumption threshold, triggering the redundant path optimization module to re-optimize. In the core scenario of a building structure encountering extreme working conditions and causing continuous collapse, the computer-aided design system for building structures endows the building structure with multi-state resilience by constructing a closed-loop generation mechanism based on dynamic graph models and multi-path optimization algorithms.

[0070] Under the premise of strictly constraining the total material consumption, the system automatically deduces a topology network with multi-level redundancy, enabling the structure to quickly reconstruct the force transmission path when locally damaged, demonstrating high adaptability and robustness in the face of unforeseen damage.

[0071] The methods for constructing an initial force transmission diagram model from an initial structural topology model and preset load data include:

[0072] Extract the component connection relationships from the initial structural topology model to generate the basic adjacency matrix;

[0073] Extract the component geometric parameters and material physical parameters from the initial structural topology model, and map the component geometric parameters and material physical parameters to the corresponding edges of the basic adjacency matrix to generate a weighted adjacency matrix;

[0074] Convert the preset load data into nodal force vectors;

[0075] The weighted adjacency matrix and the node force vectors are fused to construct the initial force transmission graph model.

[0076] This embodiment provides an in-depth explanation of the mathematical construction mechanism of the underlying graph model; the system extracts the component connection relationships in the initial structural topology model and generates a basic adjacency matrix, thereby establishing the skeleton of the spatial topology and ensuring that subsequent graph operations have a stable topological connectivity benchmark.

[0077] The system extracts the component geometric parameters and material physical parameters from the initial structural topology model, and maps the component geometric parameters and material physical parameters to the corresponding edges of the basic adjacency matrix to generate a weighted adjacency matrix;

[0078] Specifically, the geometric parameters of the components include the cross-sectional area and length, while the physical parameters of the materials include the elastic modulus. The system calculates the product of the cross-sectional area and the elastic modulus of the material for each edge, then divides it by the length of the component to obtain the axial tensile and compressive stiffness value of the component. This axial tensile and compressive stiffness value is then used as the initial weight value of the corresponding edge and assigned to the basic adjacency matrix, thereby generating a weighted adjacency matrix. This step aims to reduce the dimensionality of multi-dimensional physical and geometric properties such as cross-sectional area and elastic modulus and internalize them into graph theory weights through specific algebraic division and multiplication operations, ensuring that the graph model can truly reflect the initial mechanical transmission potential of the physical structure.

[0079] The system converts preset load data into node force vectors to characterize the excitation potential energy of the external environment. The system fuses the weighted adjacency matrix with the node force vectors to construct an initial force transmission graph model. Specifically, the system uses the weighted adjacency matrix as the input of the edge feature and spatial topology matrix of the graph model, and uses the node force vectors aligned by node number as the input of the initial node feature matrix of the graph model. By jointly encapsulating the above two mathematical matrices, the complex three-dimensional spatial entity of the building is completely transformed into a discrete mathematical graph that can be efficiently processed in parallel by a computer.

[0080] In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures effectively eliminates the dynamic mismatch problem between traditional geometric modeling and mechanical analysis by mapping the physical component properties and external load excitations to a graph neural network architecture.

[0081] This combination of features successfully constructed a high-fidelity mathematical foundation, enabling subsequent heuristic algorithms to perform high-speed parallel computation without losing key physical information, significantly improving the timeliness of early data processing.

[0082] Methods for evaluating the importance of each node in the initial force graph model using heuristic dimensionality reduction algorithms and identifying the set of high-risk nodes based on importance include:

[0083] Traverse all nodes in the initial force transmission diagram model to generate a node failure combination space;

[0084] The dimensionality of the node failure combination space is reduced based on the preset node degree and connectivity pruning rules to obtain a subset of candidate failure nodes;

[0085] Each node in the candidate failure node subset is sequentially set to the failure state, and the global stiffness decay value of the initial force transmission diagram model is calculated each time it is set to the failure state.

[0086] The global stiffness attenuation value is used as the importance of the corresponding node;

[0087] Nodes with importance greater than a preset importance threshold are extracted and combined into a high-risk node set.

[0088] This embodiment further reveals the core processing logic for overcoming the dimensional disaster caused by node failure combinations; the system traverses all nodes in the initial force transmission diagram model to generate a node failure combination space, so as to fully traverse and encompass all potential fracture risk points;

[0089] To avoid the computational bottleneck caused by full computation, the system performs dimensionality reduction on the node failure combination space based on preset node degree and connectivity pruning rules to obtain a subset of candidate failure nodes. Specifically, the preset node degree pruning rule refers to extracting the number of associated edges connecting each node in the node failure combination space as the node degree, and removing minor nodes with a degree less than or equal to a preset degree threshold, such as cantilever nodes with a degree of 1 or 2, from the combination space. The preset degree threshold is an empirical constant determined based on the basic construction requirements of statically determinate structures in the building structure design code, and is usually set to 2.

[0090] The connectivity pruning rule refers to using a preset depth-first search algorithm to calculate the number of connected components in the graph model. If the number of connected components in the graph model calculated by the depth-first search algorithm remains unchanged after a node is virtually removed and the global force transmission backbone is not severed, then the node is determined to be a non-critical connected node and is removed. The critical force-bearing nodes retained after the above dual screening of degree and connectivity constitute a subset of candidate failed nodes. This step achieves exponential compression of the computational scale by removing secondary nodes with extremely low connectivity that do not affect global connectivity.

[0091] The system sequentially sets each node in the candidate failure node subset as a failure state and calculates the global stiffness attenuation value of the initial force transmission diagram model each time it is set as a failure state. Specifically, the system extracts the initial global stiffness matrix of the initial force transmission diagram model when no node failure occurs and calculates the first eigenvalue of the initial global stiffness matrix. Specifically, the first eigenvalue is the minimum eigenvalue of the initial global stiffness matrix, which is used to characterize the weakest deformation resistance when the structure is undamaged.

[0092] After virtually removing the nodes currently set to a failed state and their associated edges, the current global stiffness matrix of the remaining structure is reassembled, and the second eigenvalue of the current global stiffness matrix is ​​calculated. Similarly, the second eigenvalue is the minimum eigenvalue of the current global stiffness matrix.

[0093] The difference between the first eigenvalue and the second eigenvalue is calculated, and the difference is divided by the first eigenvalue for normalization to obtain the global stiffness attenuation value. This attenuation value, through rigorous matrix eigenvalue comparison, intuitively quantifies the impact intensity of a single node stripping on the overall mechanical network. Based on this, the system uses the global stiffness attenuation value as the importance of the corresponding node.

[0094] The system extracts nodes with importance greater than a preset importance threshold and combines them into a high-risk node set, defining the core intervention boundary for subsequent targeted redundancy design. The preset importance threshold is dynamically calculated by the system based on the statistical distribution of importance of all nodes in the candidate failure node subset. Specifically, the system calculates the arithmetic mean and standard deviation of the importance of all candidate failure nodes, and uses the sum of the arithmetic mean and a preset multiple, such as 1.5 to 2.0 times the standard deviation, as the preset importance threshold. This accurately identifies the core nodes that have the most significant impact on global stiffness, avoiding the blindness of setting fixed thresholds based on human experience.

[0095] In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures uses preset node degrees and connectivity pruning rules to compress the originally exponentially increasing number of failure scenarios in a geometric manner, while ensuring that no core load-bearing components are omitted. This heuristic dimensionality reduction mechanism greatly improves the system's computational efficiency and real-time response capability when dealing with complex and large-scale building models, demonstrating strong practical engineering benefits.

[0096] Methods for calculating the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target node include:

[0097] Delete the target node and the associated edges connected to the target node in the initial force transmission graph model to obtain the remaining topology;

[0098] The load is redistributed to the remaining topology to obtain the initial unbalanced force vector;

[0099] Based on the initial unbalanced force vector, an incremental iterative algorithm is used to perform large deformation analysis on the remaining topology to obtain the deformation displacement data of each remaining component in the remaining topology.

[0100] Update the spatial coordinates of the remaining topology based on deformation displacement data to generate dynamic boundary conditions;

[0101] Based on dynamic boundary conditions and deformation displacement data, the internal force values ​​of each remaining component are calculated, and these internal force values ​​are used as nonlinear stress distribution data.

[0102] This embodiment details the in-situ calculation logic for nonlinear stress mutation under local failure state; the system deletes the target node and the associated edge connected to the target node in the initial force transmission diagram model to obtain the remaining topology, thereby simulating the initial state of component fracture in the real physical world.

[0103] In response to the disruption of the original force balance, the system redistributes the load on the remaining topology and obtains the initial unbalanced force vector. Specifically, the system extracts the original internal force distribution of the target node and the associated edges connected to the target node before they are removed, and applies them in reverse as equivalent nodal forces to the remaining nodes connected to the fracture in the remaining topology. The equivalent nodal forces are then superimposed with the nodal force vector converted from the original external load to calculate the resultant force vector. This resultant force vector is used as the initial unbalanced force vector, which accurately represents the transient impact force of load transfer at the moment of fracture.

[0104] Based on the initial unbalanced force vector, the system uses an incremental iterative algorithm to perform large deformation analysis on the remaining topology and obtain the deformation displacement data of each remaining component in the remaining topology. Specifically, the system divides the initial unbalanced force vector into multiple load increment steps. In each load increment step, the tangent stiffness matrix is ​​updated based on the current geometric position of the remaining component, and the Newton-Raphson iterative method is used to solve the displacement increment under that increment step until the L2 norm of the unbalanced force is less than the preset unbalanced force convergence threshold, such as 0.1% of the L2 norm of the initial unbalanced force in that increment step, or the number of iterations in the current increment step reaches the preset maximum number of iterations, such as 50 times, when convergence is determined, and the deformation displacement data is accumulated. This approach abandons the limitations of the traditional linear elastic small deformation assumption and effectively prevents the iteration process from falling into an infinite loop.

[0105] Based on this, the system updates the spatial coordinates of the remaining topology based on deformation displacement data and generates dynamic boundary conditions. Based on the dynamic boundary conditions and deformation displacement data, the system calculates the internal force values ​​of each remaining component and uses the internal force values ​​as nonlinear stress distribution data to realistically restore the mechanical characteristics under the collapse critical state.

[0106] In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures captures large spatial geometric deformations caused by component fracture in real time through incremental iterative algorithms and constructs dynamic boundary conditions in situ. This scheme realistically reproduces the nonlinear stress redistribution characteristics under the critical state of progressive collapse, making the subsequent optimization benchmark closer to the real physical collapse process and greatly improving the reliability of the analysis results.

[0107] Methods for generating secondary force transmission topology networks include:

[0108] Nodes with stress values ​​greater than a preset stress threshold in the nonlinear stress distribution data are used as source nodes, and nodes with stress values ​​less than a preset safety threshold are used as sink nodes. An optimization space is constructed in the remaining topology.

[0109] Initialize a preset number of virtual force transmission paths in the optimization space and assign an initial flow rate to each virtual force transmission path;

[0110] Calculate the three-dimensional spatial coordinates of each virtual force transmission path, and perform collision detection between the three-dimensional spatial coordinates and the preset building functional space model;

[0111] If a collision occurs, a three-dimensional geometric interference penalty constraint is applied to the virtual force transmission path that is in the collision, reducing the flow conduction rate of the corresponding virtual force transmission path; if no collision occurs, the flow conduction rate of the corresponding virtual force transmission path is maintained.

[0112] Based on dynamic boundary conditions, flow conduction rate, and preset material consumption threshold, the flow distribution of all virtual force transmission paths is dynamically updated, and virtual force transmission paths with flow exceeding the preset flow threshold are retained to generate a secondary force transmission topology network.

[0113] This embodiment reveals the core evolution process of generating secondary force transmission topology by drawing on the multi-routing mechanism of slime mold networks. The system uses nodes with stress values ​​greater than a preset stress threshold in the nonlinear stress distribution data as source nodes and nodes with stress values ​​less than a preset safety threshold as sink nodes. It constructs an optimization space in the remaining topology structure and clarifies the potential energy gradient direction of load diversion. In this process, the preset stress threshold is determined by multiplying the yield strength of the component material by a preset danger coefficient, such as 0.85 to 0.95, and is used to accurately locate dangerous areas in the plastic development stage that urgently need to be unloaded. The preset safety threshold is determined by multiplying the allowable stress of the component material by a preset bearing margin coefficient, such as 0.3 to 0.5, and is used to screen safe areas with sufficient bearing margin to receive transferred loads, thereby ensuring the mechanical rationality of the matching of source and sink nodes.

[0114] The system initializes a preset number of virtual force transmission paths in the optimization space and assigns an initial flow to each virtual force transmission path. At the same time, in order to make the virtual force transmission paths have a computable physical basis, the system assigns an initial cross-sectional area and material elastic modulus to each virtual force transmission path based on the average cross-sectional area and average material elastic modulus of all remaining components in the remaining topology, so as to start multi-path concurrent exploration.

[0115] The system calculates the three-dimensional spatial coordinates of each virtual force transmission path and performs collision detection with the preset building functional space model to ensure that the path generated by the algorithm is feasible for engineering. Specifically, the preset building functional space model refers to a set of three-dimensional bounding boxes that include the non-structural physical occupancy such as the reserved mechanical and electrical pipeline corridors, elevator shafts, HVAC equipment spaces, and personnel activity clearances inside the building. The system uses a bounding box intersection test algorithm to calculate whether the three-dimensional spatial coordinates of each virtual force transmission path intersect with the above three-dimensional bounding box set through geometric interference, thereby determining whether a collision has occurred.

[0116] Based on this, in response to a collision, the system applies a three-dimensional geometric interference penalty constraint to the virtual force transmission path that has collided, thereby reducing the flow conduction rate of the corresponding virtual force transmission path; in response to no collision, the flow conduction rate of the corresponding virtual force transmission path is maintained.

[0117] Based on dynamic boundary conditions, flow conductivity, and preset material consumption thresholds, the system dynamically updates the flow distribution of all virtual force transmission paths, retaining virtual force transmission paths with flow exceeding the preset flow threshold to generate a secondary force transmission topology network. To achieve dynamic flow updates, the system constructs a path flow evolution model, as shown in the following formula:

[0118]

[0119] in, The virtual path traffic updated in the current iteration step is dynamically calculated by the system. : Initial flow of the previous iteration step, memory storage of the previous iteration; : Dimensionless flow rate attenuation coefficient, dynamically assigned a value by the system based on the material consumption process; specifically, during the normal evolution phase, it is set as follows: Preserve historical traffic without loss;

[0120] When the total equivalent material consumption of all virtual power transmission paths in the current network approaches a preset material consumption threshold, for virtual power transmission paths whose flow rate in the previous iteration was lower than a preset proportion of the global average flow rate, the dimensionless flow rate attenuation coefficient of the corresponding path will be adjusted. The dynamic assignment value is a preset empirical constant less than one, thereby realizing a mathematical closed loop to accelerate the elimination of inefficient path decay. The flow evolution step size coefficient is pre-calibrated by the system based on the overall stiffness of the remaining topology. The calculation formula is as follows:

[0121]

[0122] in, The normalized dimensionless overall stiffness of the remaining topology relative to the initial force diagram model is calculated using the following formula:

[0123]

[0124] in, The first eigenvalue of the initial global stiffness matrix. This is the second eigenvalue of the current global stiffness matrix;

[0125] The baseline flow rate evolution constant is preset by the system. Specifically, the baseline flow rate evolution constant is positively correlated with the initial average load of the nodes in the initial force transmission diagram model, and its value is usually set to 1% to 5% of the initial average load of the nodes. : Dimensionless flow conductivity is calculated by extracting the product of the cross-sectional area of ​​the corresponding virtual force transmission path and the elastic modulus of the material, and then normalizing it by dividing it by the maximum value of the product of the cross-sectional area and elastic modulus of all components in the initial force transmission diagram model. : Dimensionless three-dimensional geometric interference penalty constraint coefficient, dynamically assigned by the collision detection module;

[0126] Specifically, when a collision occurs, the collision volume is calculated. With virtual force transmission path volume The ratio of the overlap ratio

[0127]

[0128] The dimensionless three-dimensional geometric interference penalty constraint coefficient is calculated based on the exponential mapping function. The specific formula is as follows:

[0129]

[0130] in, The preset interference sensitivity adjustment constant ranges from 3.0 to 5.0; when no collision occurs, let... ;

[0131] It should be further explained that the dynamic boundary conditions and the preset material consumption threshold play a role in global dynamic constraints and corrections in this path flow evolution model; specifically, the system uses the structural deformation state reflected in real time by the dynamic boundary conditions to adjust the flow evolution step size coefficient. and traffic throughput Dynamic corrections are made to adaptively adjust the path connectivity in areas of large deformation.

[0132] Simultaneously, after each flow update iteration, the system calculates the equivalent total material consumption of all virtual power transmission paths in the current network. If this total material consumption approaches a preset material consumption threshold, the system automatically accelerates the flow decay rate of low-flow paths, forcing their flow to rapidly drop to zero and eliminating them. This ensures that the final generated secondary power transmission topology network converges within the material cost constraint. Furthermore, after each iteration of the path flow evolution model, the system calculates the rate of change of the total network flow between two adjacent iterations. The specific method for obtaining the rate of change of the total network flow is as follows: calculate the absolute value of the difference between the total flow of all virtual power transmission paths in the current iteration and the total flow of the previous iteration, and divide this absolute value by the total flow of the previous iteration. If the rate of change of the total flow is continuous... Next, among them Less than the preset convergence tolerance threshold If the number of iterations reaches the preset maximum number of evolution generations, the flow distribution is determined to have reached a steady state, and the iterative update process is terminated.

[0133] In addition, the preset flow threshold is dynamically calculated by the system based on the initial average load-bearing flow of all components in the initial force transmission diagram model multiplied by a preset empirical reduction factor, such as 0.05 to 0.15. This is used as a cutoff boundary to filter out minor, inefficient, and redundant branches that contribute very little to the global force transmission. In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures achieves autonomous growth of multiple paths by introducing source-sink nodes and flow conduction mechanisms.

[0134] More importantly, through a rigorous three-dimensional geometric interference collision detection and penalty mechanism, it is ensured that the complex and redundant topology evolved by the computer will not conflict with the original electromechanical pipelines of the building in actual construction, thus breaking down the barrier from pure topology algorithm to actual engineering implementation.

[0135] Methods for calculating the load redistribution response rate of a secondary force transmission topology include:

[0136] In the secondary force transmission topology network, obtain the transferred load amount that is channeled from the source node to the sink node;

[0137] Obtain the original load that the target node was subjected to before it was removed;

[0138] Calculate the ratio of the transferred load to the original load to obtain the foundation diversion ratio;

[0139] Record the response time consumed by the redundant path optimization module in generating the secondary force transmission topology network;

[0140] The response time is transformed based on a preset time penalty function that monotonically decreases with the response time to obtain the time decay coefficient;

[0141] The load redistribution response rate is obtained by multiplying the basic diversion ratio by the time decay coefficient.

[0142] This embodiment clarifies a quantitative evaluation method for the system's response efficiency to sudden failures; in the secondary force transmission topology network, the system obtains the amount of transferred load from the source node to the sink node to evaluate the actual conductivity of the network;

[0143] The system obtains the original load amount borne by the target node before it is removed, as a benchmark reference; the system calculates the ratio of the transferred load amount to the original load amount to obtain the basic diversion ratio, which reflects the static load transfer integrity; based on this, in order to introduce dynamic time-effect assessment, the system records the response time consumed by the redundant path optimization module in generating the secondary force transmission topology network, and transforms the response time based on a preset time penalty function that monotonically decreases with the response time to obtain the time decay coefficient.

[0144] The system multiplies the basic diversion ratio by the time decay coefficient to obtain the load redistribution response rate. To balance the diversion volume and computational efficiency, the system defines a response rate evaluation model, as follows:

[0145]

[0146] in, : Dimensionless load redistribution response rate, calculated and output by the system in real time; The transferred load is the sum of the L2 norms of all equivalent transferred node force vectors diverted from the source node to the sink node, and is obtained from real-time statistics by the system. The original load is the sum of the L2 norms of all original node force vectors borne by the target node before it is removed. It is extracted from the initial force transmission diagram model to transform the mechanical vectors in three-dimensional space into scalar values ​​that can be subjected to algebraic division. The base of the natural logarithm is approximately 2.71828. Response time is recorded in real time by the system timer; The time penalty factor, preset according to the baseline performance of the computing platform, has the dimension of the reciprocal of time, in order to ensure the exponential term... It is a dimensionless pure number;

[0147] As an example, for a computing power platform at the level of a conventional engineering structure calculation workstation, the typical range of the time penalty factor is set to 0.01 to 0.1; the better the computing power performance of the hardware platform, the smaller the corresponding preset time penalty factor value, so as to strictly adapt to the shorter benchmark response expectation.

[0148] In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures introduces a penalty mechanism that monotonically decreases with response time, forcing the algorithm to take convergence speed into account when proposing secondary paths. This mechanism effectively avoids generating invalid and redundant solutions with ideal stress states but excessive computation time, which cannot meet the rapid iteration requirements of actual engineering projects, thus ensuring the agility of design scheduling.

[0149] Example 2:

[0150] Methods for calculating the topology redundancy index of a secondary force transmission topology include:

[0151] Extract the total number of independent force transmission paths in the secondary force transmission topology network;

[0152] Calculate the variance of the degree distribution of all nodes in the secondary force transmission topology network;

[0153] Calculate the reciprocal of the variance of the distribution to obtain the uniformity of the distribution;

[0154] The total number of independent force transmission paths and the uniformity of their distribution are assigned preset weights and then summed to obtain the redundancy index of the topology network.

[0155] This embodiment further illustrates the multidimensional calculation logic of the topology network redundancy index; the system extracts the total number of independent force transmission paths in the secondary force transmission topology network. This value intuitively reflects the absolute number of macroscopic redundant channels and is the basic redundancy guarantee against continuous structural damage; the system delves into the microscopic topology level to calculate the degree distribution variance of all nodes in the secondary force transmission topology network in order to evaluate the degree of dispersion of path connections.

[0156] Specifically, the system counts the total number of associated edges of each node in the secondary force transmission topology network as the node degree, calculates the arithmetic mean of the degrees of all nodes, and takes the average of the sum of squares of the differences between the degree of each node and the arithmetic mean to obtain the degree distribution variance. If the variance approaches zero, it indicates that the number of path connections shared by each node is relatively consistent.

[0157] The system adds a preset minimum positive real number to the variance of the degree distribution, for example... As a correction of the degree distribution variance, to prevent program crashes caused by an absolutely uniform degree distribution with a denominator of zero, the reciprocal of the correction of the degree distribution variance is calculated to obtain the distribution uniformity. This index is intended to punish fragile network structures that have excessively concentrated paths on a few hub nodes.

[0158] The system assigns preset weights to the total number of independent force transmission paths and their distribution uniformity, and then performs a weighted summation to obtain a comprehensive topology network redundancy index. The preset weights are dynamically assigned using the system's built-in hierarchical analysis algorithm combined with the building structure's safety level requirements. Specifically, the system presets a relative importance scale for the total number of independent force transmission paths and their distribution uniformity under different safety levels, constructing... The pairwise comparison matrix; for example, when the building structure is at level one safety, the distribution uniformity is set to be absolutely more important than the total number of independent force transmission paths, and the corresponding comparison matrix elements are... , ;

[0159] By calculating the normalized eigenvector corresponding to the largest eigenvalue of the matrix, the preset weights of the total number of independent force transmission paths and the distribution uniformity are obtained and assigned. In the specific calculation example of the comparison matrix of the first-level safety level above, the preset weight corresponding to the total number of independent force transmission paths is approximately 0.167, and the preset weight corresponding to the distribution uniformity is approximately 0.833.

[0160] It should be noted that, considering the significant difference in magnitude between the total number of independent force transmission paths and the distribution uniformity, the system will perform dimensionless normalization on both before performing the weighted summation operation, using preset baseline path count and baseline uniformity. That is, the total number of independent force transmission paths is divided by the baseline path count, and the distribution uniformity is divided by the baseline uniformity. Then, the normalized values ​​are multiplied by the corresponding preset weights and summed. This ensures that the contribution of each indicator to the final redundancy index is strictly matched with its weight setting, avoiding evaluation distortion caused by the absolute value of a single indicator being too large and overshadowing other indicators.

[0161] In the context of progressive collapse assessment of building structures, the computer-aided design system innovatively incorporates the reciprocal of the degree distribution variance into the redundancy assessment system. This combination of features not only requires a sufficient number of alternative paths but also mandates that these paths be evenly distributed in space. This effectively prevents the risk of secondary chain damage caused by excessive concentration of redundant paths in a local area, giving the structure extremely strong resilience against multi-point concurrent damage.

[0162] Example 3:

[0163] Methods for calculating the consumable resilience benefit ratio of secondary force transmission topologies include:

[0164] Calculate the total material volume of the secondary force transmission topology network and obtain the initial material volume of the initial force transmission diagram model;

[0165] Calculate the difference between the total material volume and the initial material volume to obtain the material increment;

[0166] Perform ultimate bearing capacity analysis on the secondary force transmission topology network to obtain the network's ultimate bearing capacity;

[0167] Obtain the initial ultimate bearing capacity of the initial force transmission diagram model;

[0168] Calculate the difference between the network's ultimate bearing capacity and the initial ultimate bearing capacity to obtain the increase in bearing capacity;

[0169] Divide the increase in load-bearing capacity by the increase in material quantity to obtain the material toughness benefit ratio.

[0170] This embodiment defines the calculation method for the material toughness benefit ratio, thereby constructing a rigid constraint boundary for the financial feasibility of the project; the system calculates the total material volume of the secondary force transmission topology network and obtains the initial material volume of the initial force transmission diagram model;

[0171] The system calculates the difference between the total material volume and the initial material volume to obtain the material increment, thus quantifying the cost change brought about by the optimization process. Based on this, the system determines whether the material increment is less than or equal to zero: if the material increment is less than or equal to zero, it indicates that the network has achieved material reduction or zero consumption growth during topology optimization, possessing a significant advantage in engineering cost. In this case, the system directly assigns a preset maximum positive real number to the material resilience benefit ratio, for example... The system terminates the subsequent ultimate bearing capacity analysis and division calculation process, and directly retains it as the superior solution; in response to the material increment being greater than zero, the system continues to execute subsequent operations;

[0172] The system performs ultimate bearing capacity analysis on the secondary force transmission topology network. Specifically, it reuses the incremental iterative large deformation analysis framework based on the Newton-Raphson iterative method mentioned above. By progressively increasing the external equivalent nodal force loads, the system obtains the ultimate bearing capacity of the network until the current overall stiffness matrix of the secondary force transmission topology network exhibits singularity or the key remaining components reach the ultimate strain state.

[0173] Simultaneously, the initial ultimate bearing capacity of the initial force transmission diagram model is obtained; based on this, the system calculates the difference between the network ultimate bearing capacity and the initial ultimate bearing capacity to obtain the bearing capacity increase, which represents a substantial enhancement of the structural safety margin.

[0174] The system divides the increase in load-bearing capacity by the increase in material quantity to obtain the material toughness benefit ratio, unifying the mechanical benefits and financial costs into a single scalar. To avoid program crashes due to division by zero or floating-point overflows caused by extremely small material quantities during topology optimization, the system adds a preset, extremely small positive real number to the material quantity before performing the division operation, for example... As a correction material increment, the increase in load-bearing capacity is divided by the correction material increment to obtain the material toughness benefit ratio, thereby ensuring the robustness of the underlying evaluation algorithm under any extreme working conditions.

[0175] In the scenario of progressive collapse assessment of building structures, the computer-aided design system effectively curbs the unreasonable tendency of topology optimization algorithms to exchange safety for unlimited material stacking by establishing a ratio model of the increase in bearing capacity to the increase in material quantity. This benefit ratio forcing system maximizes structural toughness within the most stringent material cost boundary by reorganizing the underlying topology rather than simply increasing material quantity without constraints, thus ensuring the financial feasibility of the generated scheme in actual construction.

[0176] Example 4:

[0177] Methods for adjusting the 3D geometric interference penalty constraint or preset material consumption threshold, and triggering the redundant path optimization module to re-perform multi-path optimization, include:

[0178] If the load redistribution response rate is less than the preset response rate threshold, a first adjustment coefficient less than one is obtained, the three-dimensional geometric interference penalty constraint is multiplied by the first adjustment coefficient for relaxation, and the redundant path optimization module is triggered.

[0179] If the load redistribution response rate is greater than or equal to the preset response rate threshold and the topology network redundancy index is less than the preset redundancy threshold, then a second adjustment coefficient greater than one is obtained, the three-dimensional geometric interference penalty constraint is multiplied by the second adjustment coefficient for tightening, and the redundant path optimization module is triggered.

[0180] If the load redistribution response rate is greater than or equal to the preset response rate threshold, the topology network redundancy index is greater than or equal to the preset redundancy threshold, and the material resilience benefit ratio is less than the preset benefit ratio threshold, then the preset material reduction step size is obtained, the preset material consumption threshold is reduced based on the preset material reduction step size, and the redundant path optimization module is triggered.

[0181] This embodiment reveals the adaptive closed-loop feedback adjustment mechanism of the system when facing a substandard topology network; in response to the load redistribution response rate being less than the preset response rate threshold, the system determines that the current optimization is blocked and the load cannot be effectively transferred. At this time, the system obtains a first adjustment coefficient less than one, multiplies the dimensionless three-dimensional geometric interference penalty constraint coefficient by the first adjustment coefficient for relaxation, and triggers the redundant path optimization module to prioritize the preservation of the main force transmission path of the structure by moderately tolerating slight spatial interference.

[0182] If the load redistribution response rate is greater than or equal to the preset response rate threshold and the topology network redundancy index is less than the preset redundancy threshold, the system determines that the load transfer is extremely fast but the path is too simple and lacks multi-level redundancy. It obtains a second adjustment coefficient greater than one, multiplies the dimensionless three-dimensional geometric interference penalty constraint coefficient by the second adjustment coefficient for tightening, and triggers the redundant path optimization module. The strict interference penalty will force the algorithm to abandon the single minimum geometric impedance path and instead spread outwards to evolve into a mesh of multiple branching paths.

[0183] In response to the load redistribution response rate and topology redundancy index both meeting the standards, and the material consumption toughness benefit ratio being less than the preset benefit ratio threshold, the system determines that the mechanical performance and redundancy meet the standards but the material waste is serious. Based on this, a preset material reduction step size is obtained, the preset material consumption threshold is reduced based on the preset material reduction step size, and the redundant path optimization module is triggered to force the algorithm to perform topology simplification iteration.

[0184] It should be noted that the parameters in the above feedback adjustment mechanism all have clear sources of acquisition and value boundaries; specifically, the first adjustment coefficient is an empirical constant less than one, calibrated in advance through a large number of structural interference evolution experiments, and usually takes a value of 0.8 to 0.95; the second adjustment coefficient is an empirical constant greater than one, calibrated in advance, and usually takes a value of 1.1 to 1.5; the preset material reduction step size is obtained by dynamically calculating a fixed percentage of the current preset material consumption threshold, such as 5%, thereby ensuring that the iteration process converges smoothly without violent oscillations;

[0185] Furthermore, it should be noted that the core evaluation preset thresholds involved in the resilience assessment and optimization module of this system all have clear engineering value boundaries: the preset response rate threshold is usually set to 0.85 to 0.95 to ensure the efficiency of foundation load transfer after local failure of the structure; the preset redundancy threshold is set according to the safety level of the building structure, generally ranging from 2.0 to 5.0, to quantify the safety redundancy space of multi-path force transmission; the preset benefit ratio threshold is usually set to 1.5 to 3.0 to achieve a cost balance between load-bearing capacity improvement and material consumption; and the preset material consumption threshold is usually set to 105% to 115% of the initial material volume of the initial force transmission diagram model, thereby strictly limiting the excessive increase in overall material due to the progressive collapse prevention redundancy design.

[0186] In the scenario of progressive collapse assessment of building structures, the computer-aided design system for building structures has constructed a highly adaptive cascade constraint adjustment mechanism. Based on the assessment shortcomings in different dimensions, the system can autonomously decide whether to relax spatial constraints, tighten interference penalties, or compress material thresholds, thus solving the local deadlock problem that often occurs in multi-objective optimization. This ensures that the multi-path optimization algorithm can converge stably and efficiently to the global optimal solution that takes into account toughness, space, and cost.

[0187] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A computer-aided design-based building structure design system, characterized in that, include: The initial data acquisition module is used to acquire the initial structural topology model and the preset load data, and to construct an initial force transmission diagram model based on the initial structural topology model and the preset load data; The high-risk node locking module is used to simulate node failure in the initial force transmission diagram model, evaluate the importance of each node in the initial force transmission diagram model through a heuristic dimensionality reduction algorithm, and lock the set of high-risk nodes based on the importance. The nonlinear response analysis module is used to select the nodes in the high-risk node set as target nodes and remove them from the initial force transmission diagram model, and to calculate the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target nodes. The redundant path optimization module is used to perform multi-path optimization based on the nonlinear stress distribution data and the dynamic boundary conditions, under the preset material consumption threshold and three-dimensional geometric interference penalty constraints, to generate a secondary force transmission topology network. The resilience assessment and optimization module is used to calculate the load redistribution response rate, topology redundancy index, and material resilience benefit ratio of the secondary force transmission topology network; compare the load redistribution response rate, topology redundancy index, and material resilience benefit ratio with preset response rate thresholds, preset redundancy thresholds, and preset benefit ratio thresholds, respectively; if all are greater than or equal to the corresponding preset thresholds, the secondary force transmission topology network is output as the target structural topology network; otherwise, the three-dimensional geometric interference penalty constraint or the preset material consumption threshold is adjusted, and the redundant path optimization module is triggered to re-perform multi-path optimization.

2. The computer-aided design system for building structures according to claim 1, characterized in that, The method for constructing an initial force transmission diagram model based on the initial structural topology model and the preset load data includes: Extract the component connection relationships from the initial structural topology model to generate a basic adjacency matrix; Extract the component geometric parameters and material physical parameters from the initial structural topology model, and map the component geometric parameters and material physical parameters to the corresponding edges of the basic adjacency matrix to generate a weighted adjacency matrix; The preset load data is converted into nodal force vectors; The weighted adjacency matrix and the node force vector are fused to construct the initial force transmission graph model.

3. The computer-aided design-based building structure design system according to claim 2, characterized in that, The method for evaluating the importance of each node in the initial force graph model using a heuristic dimensionality reduction algorithm, and then identifying a set of high-risk nodes based on the importance, includes: Traverse all nodes in the initial force transmission diagram model to generate a node failure combination space; The dimensionality of the node failure combination space is reduced based on preset node degree and connectivity pruning rules to obtain a subset of candidate failure nodes. Each node in the candidate failure node subset is sequentially set to a failure state, and the global stiffness attenuation value of the initial force transmission diagram model is calculated each time it is set to a failure state. The global stiffness attenuation value is used as the importance of the corresponding node; Nodes with an importance greater than a preset importance threshold are extracted and combined into the high-risk node set.

4. The computer-aided design system for building structures according to claim 3, characterized in that, The methods for calculating the nonlinear stress distribution data and dynamic boundary conditions of the remaining topology after removing the target node include: The target node and the associated edges connected to the target node are deleted from the initial force transmission graph model to obtain the remaining topology. The load is redistributed to the remaining topology to obtain the initial unbalanced force vector; Based on the initial unbalanced force vector, an incremental iterative algorithm is used to perform large deformation analysis on the remaining topology to obtain the deformation displacement data of each remaining component in the remaining topology. The spatial coordinates of the remaining topology are updated based on the deformation displacement data to generate the dynamic boundary conditions. Based on the dynamic boundary conditions and the deformation displacement data, the internal force values ​​of each remaining component are calculated, and the internal force values ​​are used as the nonlinear stress distribution data.

5. The computer-aided design system for building structures according to claim 4, characterized in that, Methods for generating secondary force transmission topology networks include: Using nodes in the nonlinear stress distribution data whose stress values ​​are greater than a preset stress threshold as source nodes and nodes whose stress values ​​are less than a preset safety threshold as sink nodes, an optimization space is constructed in the remaining topology. A preset number of virtual force transmission paths are initialized in the optimization space, and each virtual force transmission path is assigned an initial flow rate. Calculate the three-dimensional spatial coordinates of each virtual force transmission path, and perform collision detection between the three-dimensional spatial coordinates and the preset building functional space model; If a collision occurs, the three-dimensional geometric interference penalty constraint is applied to the virtual force transmission path where the collision occurs, reducing the flow conduction rate of the corresponding virtual force transmission path; if no collision occurs, the flow conduction rate of the corresponding virtual force transmission path is maintained. Based on the dynamic boundary conditions, the flow conduction rate, and the preset material consumption threshold, the flow distribution of all virtual force transmission paths is dynamically updated, and virtual force transmission paths with flow rates greater than the preset flow rate threshold are retained to generate the secondary force transmission topology network.

6. The computer-aided design system for building structures according to claim 5, characterized in that, The method for calculating the load redistribution response rate of the secondary force transmission topology network includes: In the secondary force transmission topology network, the amount of transferred load from the source node to the sink node is obtained; Obtain the original load that the target node was subjected to before it was removed; Calculate the ratio of the transferred load to the original load to obtain the foundation diversion ratio; Record the response time consumed by the redundant path optimization module in generating the secondary force transmission topology network; The response time is transformed based on a preset time penalty function that monotonically decreases with the response time to obtain a time decay coefficient; The load redistribution response rate is obtained by multiplying the basic diversion ratio by the time decay coefficient.

7. The computer-aided design system for building structures according to claim 6, characterized in that, The method for calculating the topology redundancy index of the secondary force transmission topology network includes: Extract the total number of independent force transmission paths in the secondary force transmission topology network; Calculate the degree distribution variance of all nodes in the secondary force transmission topology network; Calculate the reciprocal of the variance of the degree distribution to obtain the distribution uniformity; The total number of independent force transmission paths and the distribution uniformity are assigned preset weights and then summed to obtain the redundancy index of the topology network.

8. The computer-aided design system for building structures according to claim 7, characterized in that, The method for calculating the consumable toughness benefit ratio of the secondary force transmission topology network includes: Calculate the total material volume of the secondary force transmission topology network and obtain the initial material volume of the initial force transmission diagram model; Calculate the difference between the total material volume and the initial material volume to obtain the material increment; Perform ultimate bearing capacity analysis on the secondary force transmission topology network to obtain the network's ultimate bearing capacity; Obtain the initial ultimate bearing capacity of the initial force transmission diagram model; Calculate the difference between the network's ultimate bearing capacity and the initial ultimate bearing capacity to obtain the increase in bearing capacity; Divide the increase in load-bearing capacity by the increase in material quantity to obtain the toughness benefit ratio of the consumable.

9. The computer-aided design system for building structures according to claim 8, characterized in that, The method of adjusting the three-dimensional geometric interference penalty constraint or the preset material consumption threshold, and triggering the redundant path optimization module to re-perform multi-path optimization, includes: If the load redistribution response rate is greater than or equal to the preset response rate threshold, and the topology network redundancy index is less than the preset redundancy threshold, then a second adjustment coefficient greater than one is obtained, the constraint coefficient corresponding to the three-dimensional geometric interference penalty constraint is multiplied by the second adjustment coefficient for tightening, and the redundant path optimization module is triggered. If the load redistribution response rate is greater than or equal to the preset response rate threshold, the topology network redundancy index is greater than or equal to the preset redundancy threshold, and the material resilience benefit ratio is less than the preset benefit ratio threshold, then a preset material reduction step size is obtained, the preset material consumption threshold is reduced based on the preset material reduction step size, and the redundant path optimization module is triggered.

Images